Weight regain after stopping GLP-1 treatment is a substantial average finding in several withdrawal studies, but it is not a prediction that everyone will regain every kilogram. Results depend on the medicine, study population, treatment period and what support or treatment continued after withdrawal.
The evidence also distinguishes stopping completely from continuing at a reduced dose or switching medicines. With maintenance research published in 2026, those categories should remain separate when interpreting a headline or an account of someone’s experience.
Weight regain after stopping GLP-1: the key studies
| Study | What happened after initial treatment? | Main observation relevant to withdrawal |
|---|---|---|
| STEP 1 extension | Semaglutide and the study lifestyle intervention ended; one-year follow-up | About two-thirds of the preceding weight loss was regained on average in the former semaglutide group |
| STEP 4 | Participants continued semaglutide or switched to placebo; lifestyle intervention continued | From randomization, mean weight changed by −7.9% with continuation versus +6.9% after the switch |
| SURMOUNT-4 | Participants continued tirzepatide or switched to placebo | From randomization, mean weight changed by −5.5% with continuation versus +14.0% after the switch |
| SURMOUNT-MAINTAIN, 2026 | Compared continuation, a defined dose reduction and placebo after an initial tirzepatide period | Maintenance treatment and complete withdrawal produced different results |
Original sources: STEP 1 extension, STEP 4, SURMOUNT-4 and SURMOUNT-MAINTAIN. The percentages use different reference points and should not be treated as a head-to-head comparison.
What did the STEP 1 extension show?
The exploratory extension analyzed 327 participants after the main trial. In the former semaglutide group, mean weight loss had been 17.3% at the end of treatment. Over the following year, participants regained 11.6 percentage points, leaving mean weight 5.6% below the original baseline. Read the STEP 1 extension.
That is the source of the widely repeated “two-thirds” finding. The same report states that the study lifestyle intervention ended alongside the medication, and that the extension involved a subset of trial participants. It therefore does not isolate medication withdrawal from every other change in support.
This is also why the result should not become “everyone gained it all back.” The study reports an average change in a particular group, with uncertainty and variation around that average.
What does STEP 4 add?
STEP 4 used a randomized continuation-versus-withdrawal design after a 20-week semaglutide run-in. A total of 803 participants were randomized, and lifestyle intervention continued in both groups during the following 48 weeks. The results in the table therefore address a different question from the STEP 1 off-treatment extension. See the STEP 4 randomized trial.
Keeping that design difference visible helps answer a common question: was support still provided after the study medicine stopped? It also prevents one trial’s limitations from being incorrectly applied to every withdrawal study.
What happened after tirzepatide withdrawal in SURMOUNT-4?
SURMOUNT-4 randomized 670 participants after a 36-week open-label period with mean weight reduction of 20.9%. The randomized phase lasted another 52 weeks. Participants who switched to placebo regained substantial weight on average; those who continued tirzepatide lost additional weight. Read the original SURMOUNT-4 report.
Its +14.0% result uses weight at randomization as the reference. That is different from the percentage of previously lost weight regained. A chart that omits the denominator can make the result look much smaller or larger than it is.
For the distinctions between the two molecules, see our semaglutide vs tirzepatide guide. Tirzepatide’s dual GIP/GLP-1 activity also makes “GLP-1 drugs” a convenient discussion category rather than a precise description of every molecule.
How should percentages of regain be read?
Three measures are often mixed together: change from the original starting weight, change from weight when treatment stopped, and the share of lost weight regained.
Consider an illustrative person starting at 100 kg, reaching 80 kg and later weighing 90 kg. That means:
- 10% below the original starting weight;
- 12.5% above the weight at stopping;
- 50% of the previously lost weight regained.
All three descriptions refer to the same example. These are explanatory calculations, not participant data or a forecast. Before comparing two headlines, identify which measure each uses.
What does 2026 maintenance research change?
SURMOUNT-MAINTAIN compared continued tirzepatide, a defined reduction strategy and placebo after a longer initial treatment period. The trial reported that continued treatment and reduced-dose treatment were superior to placebo for maintaining weight reduction. This adds evidence about a specific maintenance approach; it does not establish that reducing and stopping are equivalent. See the 2026 Lancet trial record and the sponsor’s accompanying results announcement.
A planned dose reduction in a study is also different from irregular use, a self-designed taper or switching to an unverified material. Evidence should follow the intervention actually studied. This article does not provide a stopping, restarting or maintenance schedule.
Why do personal accounts vary?
An individual report can describe an outcome that differs from a trial’s average without disproving the trial. It usually cannot show how common that outcome is, because the reader does not know the total number of comparable people, their follow-up or their other treatment changes.
For the same reason, a study average cannot determine exactly what will happen to one person. Neither a success story nor a mean regain percentage is enough to design someone’s care.
If access, tolerability or a change in circumstances raises the question of stopping, review it with the prescribing clinician. Useful questions include what follow-up is needed, what outcomes will be monitored and which evidence applies to the person’s circumstances.
Where laboratory research materials fit
These clinical studies do not validate catalog materials for personal maintenance use. Our peptide basics guide and research peptides vs compounded peptides explanation help distinguish molecule names from product categories.
For a laboratory project, material identity and batch documentation remain separate questions from clinical outcomes. Consult the quality and COA guide or contact Certiva about a defined research specification.
Evidence reviewed October 14, 2026. This is a research summary, not personal medical advice. Certiva materials are for laboratory research use only, not for human consumption, and are not substitutes for prescribed medicines.
Frequently asked questions
Does everyone regain all their weight after stopping GLP-1 treatment?
No. Trials report group averages and variation. Several studies found substantial average regain after withdrawal, but that does not mean every participant regained all previously lost weight.
What does the STEP 1 finding of two-thirds regained mean?
It describes the share of prior weight loss regained on average in the semaglutide extension group over one year. It does not mean a two-thirds increase in total body weight.
What does 14% regain mean in SURMOUNT-4?
It is the mean percentage weight increase from the week-36 randomization weight over the following 52 weeks in participants switched to placebo. It is not 14% of the kilograms previously lost.
Is reducing treatment the same as stopping it?
No. Reduced-dose treatment, switching to another medicine and complete withdrawal are separate interventions. The 2026 SURMOUNT-MAINTAIN trial studied a defined reduction strategy, not every possible tapering schedule.
Do these trials justify using research peptides for maintenance?
No. Their results concern controlled clinical interventions. A research-use material is not a prescribed medicine, and these studies do not establish its suitability for personal maintenance treatment.
For research use only. Not for human consumption. This article is educational and makes no medical, therapeutic, or dosing claims.
← More gLP-1 & weight management guides · Have a question? Start an inquiry →